totalizer counting less

V

Thread Starter

vinod ghate

Dear All,

I have configured a Steam totalizer in citect SCADA 5.21. I have used the Steam flow indication along with steam pressure to give a compensated steam flow. & on that basis the steam is totalized. But this totalizer is counting @30/40 counts less @60000Kg/Hr. & I'm unable to tress why this is happening. Can any one help me in this regards?

Thanks in advance.
 
You're looking at an error of 5 parts in 10,000.

How do you know that the totalizer value is truly low at that level of error?

Is the total calculated from an analog value or digital pulse?

When a totalizer is truly 'slow', my experience is for digital pulse inputs is

- pulse rate is too fast for the DI input, or pulse duration is too short for the DI to recognize the pulse

- hardware scan rate intermittently slows for whatever reason, causing a missed pulse

- the occasional pulse gets lost in wireless propagation

When using analog values
- comparing an analog value from a field instrument, not the acutal analog value used by the totalizer where analog calibration differences contribute to a flow rate difference.

- when analog flow rates are used there, IEEE floating point can cause a rounding error and an
arithmetic error.

- when very small values are added to very, very large values the small value isn't 'seen' in the
sum. The work-around fix is to use two registers. For further explanations, refer to this

http://www.plctalk.net/qanda/showthread.php?t=52310
(take out any spaces the forum inserts into the URL)

- IEEE floating point doesn't recognize values less than 1 when added to very large values.

I'd suggest 'proving' the totalizer with a known flow rate signal from a source good enough to be used for calibration.

If the totalizer is computing its total from a flow rate pulse signal, then set up a channel and prove the totalizer with a pulse generator that can be set at a fixed output frequency in the same range as your flow meter's output. Run it for an extended period and see whether it does what it's supposed to.

If the totalizer is computing its total from a flow rate analog signal flow, then set up a channel and prove the totalizer using a fixed constant substituted as the analog flow rate signal. A drift free calibrator's analog output could be used, as well, but a fixed constant is not going to change under any circumstance.
 
G
I'm using analog 4-20 mA signal of steam flow & steam pressure & then using mathematical calculation in scada. i have calculated compensated steam flow. & on this basis in scada it self i'm totalizing the steam flow. Previously it was working fine. but after i made changes in software (added some loops) it is giving 30/40 counts/min. less. i have observed that it stops totalizing for 10/12 seconds. (Scan rate issues?) Steam flow span is 10000 Kg/Hr. (normal usage 6500 kg/hr.) I'm observing the totalizer when steam flow is stable (1 min.)

Also the reading for 24 hr.s is also less by 50 tons.
 
C

Chris Jennings

Why do the calculation in Citect? Can't you do this calculation in the PLC or perhaps the instrument. Unless you can guarantee scan rates (i.e. synchronous polling) the calculation result could vary.

I would try and move the calculation to the PLC and then grab the data from there with Citect.

Chris Jennings
 
Top